quantify_verbatim: Regress phrases on a quantitative variable from survey and...

In chrisumphlett/openEndedAnalyzeR: Tidy analysis of open-ended survey question responses

Description

Regresses phrases on a quantitative variable from survey and produces a plot showing the associated change in quantitative if a particular phrase is used in the verbatim.

Usage

quantify_verbatim(surveydata, tidydata, id = id, quant_var,
  min_occur = 3, min_phrases = 25, max_phrases = 80,
  pval_limit = 0.1, xreg = c(), remove_stop = TRUE)

Arguments

surveydata	Data frame with the survey data
tidydata	Data frame with the unnested phrases (usually from tidy_verbatim())
id	Unique survey id. By default, can be omitted if column is named 'id'
quant_var	Numerical column in survey data frame on which phrases will be regressed
min_occur	Number of times a phrase must occur to be included
min_phrases	Force the plot to include at least this many phrases
max_phrases	Maximum number of phrases to show in plot
pval_limit	Threshold from 0 - 1. Phrases below the threshold will be highlighted on the plot. Default is 0.1
xreg	Optional list of columns from survey data frame to be used as additional control variables in regression
remove_stop	Do you want to remove common "stop-words"? Applies only to single-word phrases ("unigrams"). If so, use TRUE

chrisumphlett/openEndedAnalyzeR documentation built on Dec. 25, 2019, 4:10 p.m.